Quality, Safety and Sustainability Value Chain Analysis of Fruit Exports: A Case Study of Thai Fresh Mangosteen exported to China

Director(s): 
Co-supervisor(s): 
External supervisors: 

G. NEUBERT

Starting date: 
October 2016
Defense date: 
Thursday 09 July 2020
Host institution: 

Purpose: Adapting the distribution of fresh fruit to meet consumers' specifications and requirements for quality, safety and sustainability while controlling costs is a complex challenge. Fruit Distribution Management (FDM) is characterized by long-distance transportation, continuous decline in fruit quality in various form combined with significant supply, demand, and price uncertainties, and relatively thin margins. For these reasons, Fruit Distribution Management needs modern decision-making tools to be effective. The main objective of this research is to propose a method for optimizing the logistics needed to export fresh fruit which takes into account the specific challenges of this kind of product.

Design/methodology/approach: The methodology is divided into 3 phases.

First, the input factors that constitute the logistics costs and determine the loss of quality due to mechanical injuries (MI) and natural decay (ND), as well as the CO2 emissions of each activity are developed. This phase is based on an extended analysis of the literature and the use of real data. Second, we have developed a multi-objective optimization model that integrates both the distribution and inventory planning of fresh fruit (DIP model). This model takes into account the logistics cost, the total greenhouse gas (GHG) emissions, as well as the quality loss due to MI and ND based on two models of kinetic shelf life: zero order (DIP). - ZO) and exponential decay (DIP-ED). The DIP model objective is to optimize the planning for the distribution and storage for fresh fruit. It is based on a bi objective function: maximizing the total net profit (NP) and the percentage of the remaining fruit quality (% QR). Finally, to address this issue, we have applied a Non-Dominated Sorting Genetic Algorithm II (NSGA-II). A numerical study allows us to analyze the solution spaces of the two kinetic shelf life models (DIP-ZO and DIP-ED) according to the variation of supply, demand, and the market price. This study also includes CO2 emissions analysis. The results based on the NSGA-II are then compared with the results of another genetic algorithm based on an adjusted weight-sum approach (AWS-GA) to evaluate the performance of the algorithm.

Findings: the results we obtained indicate the efficiency of the NSGA-II algorithm to solve the problem of fresh fruit distribution. Both optimal Pareto front curves of DIP-ZO and DIP-ED show a strong positive relationship between %RQ and NP. They also indicate a negative correlation between % RQ and CO2 emission: reduction of % RQ in both cases is correlated with an increase in CO2. The decision variables can be used to decide on the best schedule for transporting and storing fruit in order to optimize the net profit and remaining quality.

Practical implications: The proposed model (DIP-model) allows a significant improvement in the benefits related to the export of fresh fruit while minimizing the loss of quality and the environmental impact. It is a decision support tool that allows exporters to better plan the transport and storage of exported fresh fruits.

Originality/value: The distribution and inventory planning model developed in this thesis provides solutions that integrate the kinetics of degradation of fresh fruit. While the other operational research models found in the literature only focus on Natural Decay based on a zero order reaction, our model introduces mechanical injuries, natural decay (zero order and Exponential) and consumer evaluation in the optimization process. This is an innovative approach because former models do not reflect reality, in particular for the international trade in fresh fruit which requires long distance transport.

Keywords: Distribution management, Perishable fruit; Shelf life model; Multi-objective decision making; CO2 emission; NSGA II.